Merge branch 'main' into 480-assert-bad

ga4gh · Jan 31, 2025 · 89ea006 · 89ea006
2 parents 9ae00c3 + 51f08c9
commit 89ea006
Show file tree

Hide file tree

Showing 35 changed files with 2,166 additions and 1,473 deletions.
diff --git a/.github/workflows/python-cqa.yml b/.github/workflows/python-cqa.yml
@@ -2,21 +2,34 @@ name: Python CQA
 
 on: [push, pull_request]
 
-
-# TODO: separate lint and test jobs; lint first, test "needs" lint
-# https://docs.github.com/en/actions/reference/workflow-syntax-for-github-actions#jobs
-
 jobs:
-  test:
+  lint:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          submodules: recursive
+      - uses: actions/setup-python@v5
+      - name: Update pip and setuptools
+        run: |
+          python -m pip install --upgrade pip setuptools
+      - name: Install dependencies
+        run: pip install -e '.[dev]'
+      - name: Check style
+        run: python3 -m ruff check --output-format=github
+      - name: Check formatting
+        run: python3 -m ruff format --check
 
+  test:
     runs-on: ubuntu-latest
+    needs:
+      - lint
     strategy:
       fail-fast: false
       matrix:
         python-version: ["3.10", "3.11", "3.12"]
     env:
       SEQREPO_ROOT_DIR: ./tests/data/seqrepo/latest
-
     steps:
     - uses: actions/checkout@v4
       with:
@@ -43,7 +56,15 @@ jobs:
 
   precommit_hooks:
     runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        cmd:
+          - "end-of-file-fixer"
+          - "trailing-whitespace"
+          - "mixed-line-ending"
     steps:
-    - uses: actions/checkout@v4
-    - uses: actions/setup-python@v5
-    - uses: pre-commit/[email protected]
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+      - uses: pre-commit/[email protected]
+        with:
+          extra_args: ${{ matrix.cmd }} --all-files
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -11,4 +11,10 @@ repos:
         args: [ --allow-missing-credentials ]
       - id: mixed-line-ending
         args: [ --fix=lf ]
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.9.4  # ruff version
+    hooks:
+      - id: ruff-format
+      - id: ruff
+        args: [ --fix, --exit-non-zero-on-fix ]
 minimum_pre_commit_version: 4.0.1
diff --git a/Makefile b/Makefile
@@ -16,8 +16,6 @@ else
 endif
 XRM=xargs -0${_XRM_R} rm
 
-PKG=ga4gh.vrs
-PKGD=$(subst .,/,${PKG})
 PYV:=3.12
 VEDIR=venv/${PYV}
 
@@ -89,18 +87,15 @@ doctest:
 ############################################################################
 #= UTILITY TARGETS
 
-# N.B. Although code is stored in github, I use hg and hg-git on the command line
-#=> reformat: reformat code with yapf and commit
-.PHONY: reformat
-reformat:
-	@if ! git diff --cached --exit-code; then echo "Repository not clean" 1>&2; exit 1; fi
-	yapf -i -r "${PKGD}" tests
-	git commit -a -m "reformatted with yapf"
+#=> reformat: reformat code with ruff
+.PHONY: format
+format:
+	python3 -m ruff format
 
 #=> lint -- static analysis check
 .PHONY: lint
 lint:
-	pylint src/ga4gh/{core,vrs} | tee $@
+	python3 -m ruff check --fix
 
 #=> docs -- make sphinx docs
 .PHONY: docs

diff --git a/docker-compose.yml b/docker-compose.yml
@@ -1,5 +1,3 @@
-version: "3"
-
 services:
   seqrepo:
     image: biocommons/seqrepo:2024-02-20

diff --git a/pyproject.toml b/pyproject.toml
@@ -59,8 +59,7 @@ dev = [
     "pyyaml",
     # style
     "pre-commit>=4.0.1",
-    "pylint",
-    "yapf",
+    "ruff==0.9.4",
     # docs
     "sphinx",
     "sphinx_rtd_theme",
@@ -96,9 +95,139 @@ addopts = "--cov-report=term-missing --cov=ga4gh"
 testpaths = ["tests", "src"]
 doctest_optionflags = "ALLOW_UNICODE ALLOW_BYTES ELLIPSIS IGNORE_EXCEPTION_DETAIL NORMALIZE_WHITESPACE"
 
-[tool.yapf]
-based_on_style = "pep8"
-column_limit = 120
-spaces_before_comment = 4
-split_before_logical_operator = true
-split_before_named_assigns = true
+[tool.ruff]
+src = ["src"]
+exclude = ["docs", "misc", "notebooks", "submodules"]
+
+[tool.ruff.lint]
+select = [
+    "F",  # https://docs.astral.sh/ruff/rules/#pyflakes-f
+    "E", "W",  # https://docs.astral.sh/ruff/rules/#pycodestyle-e-w
+    "I",  # https://docs.astral.sh/ruff/rules/#isort-i
+    "N",  # https://docs.astral.sh/ruff/rules/#pep8-naming-n
+    "D",  # https://docs.astral.sh/ruff/rules/#pydocstyle-d
+    "UP",  # https://docs.astral.sh/ruff/rules/#pyupgrade-up
+    "ANN",  # https://docs.astral.sh/ruff/rules/#flake8-annotations-ann
+    "ASYNC",  # https://docs.astral.sh/ruff/rules/#flake8-async-async
+    "S",  # https://docs.astral.sh/ruff/rules/#flake8-bandit-s
+    "B",  # https://docs.astral.sh/ruff/rules/#flake8-bugbear-b
+    "A",  # https://docs.astral.sh/ruff/rules/#flake8-builtins-a
+    "C4",  # https://docs.astral.sh/ruff/rules/#flake8-comprehensions-c4
+    "DTZ",  # https://docs.astral.sh/ruff/rules/#flake8-datetimez-dtz
+    "T10",  # https://docs.astral.sh/ruff/rules/#flake8-datetimez-dtz
+    "EM",  # https://docs.astral.sh/ruff/rules/#flake8-errmsg-em
+    "LOG",  # https://docs.astral.sh/ruff/rules/#flake8-logging-log
+    "G",  # https://docs.astral.sh/ruff/rules/#flake8-logging-format-g
+    "INP",  # https://docs.astral.sh/ruff/rules/#flake8-no-pep420-inp
+    "PIE",  # https://docs.astral.sh/ruff/rules/#flake8-pie-pie
+    "T20",  # https://docs.astral.sh/ruff/rules/#flake8-print-t20
+    "PT",  # https://docs.astral.sh/ruff/rules/#flake8-pytest-style-pt
+    "Q",  # https://docs.astral.sh/ruff/rules/#flake8-quotes-q
+    "RSE",  # https://docs.astral.sh/ruff/rules/#flake8-raise-rse
+    "RET",  # https://docs.astral.sh/ruff/rules/#flake8-return-ret
+    "SLF",  # https://docs.astral.sh/ruff/rules/#flake8-self-slf
+    "SLOT",  # https://docs.astral.sh/ruff/rules/#flake8-slots-slot
+    "SIM",  # https://docs.astral.sh/ruff/rules/#flake8-simplify-sim
+    "ARG",  # https://docs.astral.sh/ruff/rules/#flake8-unused-arguments-arg
+    "PTH",  # https://docs.astral.sh/ruff/rules/#flake8-use-pathlib-pth
+    "PGH",  # https://docs.astral.sh/ruff/rules/#pygrep-hooks-pgh
+    "PLC",  # https://docs.astral.sh/ruff/rules/#convention-c
+    "PLE",  # https://docs.astral.sh/ruff/rules/#error-e_1
+    "TRY",  # https://docs.astral.sh/ruff/rules/#tryceratops-try
+    "PERF",  # https://docs.astral.sh/ruff/rules/#perflint-perf
+    "FURB",  # https://docs.astral.sh/ruff/rules/#refurb-furb
+    "RUF",  # https://docs.astral.sh/ruff/rules/#ruff-specific-rules-ruf
+]
+fixable = ["ALL"]
+# ANN002 - missing-type-args
+# ANN003 - missing-type-kwargs
+# D203 - one-blank-line-before-class
+# D205 - blank-line-after-summary
+# D206 - indent-with-spaces*
+# D213 - multi-line-summary-second-line
+# D300 - triple-single-quotes*
+# D400 - ends-in-period
+# D415 - ends-in-punctuation
+# E111 - indentation-with-invalid-multiple*
+# E114 - indentation-with-invalid-multiple-comment*
+# E117 - over-indented*
+# E501 - line-too-long*
+# W191 - tab-indentation*
+# S321 - suspicious-ftp-lib-usage
+# PLC0206 - dict-index-missing-items
+# *ignored for compatibility with formatter
+ignore = [
+    "ANN002", "ANN003",
+    "D203", "D205", "D206", "D213", "D300", "D400", "D415",
+    "E111", "E114", "E117", "E501",
+    "W191",
+    "S321",
+    "PLC0206",
+]
+exclude = [
+    "src/ga4gh/vrs/models.py",  # there be monsters here
+]
+
+[tool.ruff.lint.per-file-ignores]
+# ANN001 - missing-type-function-argument
+# ANN2 - missing-return-type
+# D100 - undocumented-public-module
+# D102 - undocumented-public-class
+# D103 - undocumented-public-function
+# S101 - assert
+# B011 - assert-false
+# INP001 - implicit-namespace-package
+# SLF001 - private-member-access
+"tests/*" = [
+    "ANN001",
+    "ANN2",
+    "D100",
+    "D102",
+    "D103",
+    "S101",
+    "B011",
+    "INP001",
+    "SLF001",
+]
+"src/ga4gh/vrs/utils/hgvs_tools.py" = [
+    "ANN001",
+    "ANN201",
+    "ANN202",
+    "D102",
+]
+"src/ga4gh/vrs/normalize.py" = [
+    "ANN001",
+    "ANN201",
+    "ANN202",
+]
+"src/ga4gh/vrs/extras/vcf_annotation.py" = [
+    "PTH123",  # see https://github.com/ga4gh/vrs-python/issues/482
+]
+"src/ga4gh/vrs/extras/translator.py" = [
+    "ANN001",
+    "ANN101",
+    "ANN2",
+]
+"src/ga4gh/vrs/extras/object_store.py" = [
+    "ANN",
+    "D",
+]
+"src/ga4gh/vrs/enderef.py" = [
+    "ANN",
+    "D",
+]
+"src/ga4gh/vrs/dataproxy.py" = [
+    "B019",
+]
+"src/ga4gh/core/pydantic.py" = [
+    "ANN401",
+]
+"src/ga4gh/core/models.py" = [
+    "UP007",
+]
+
+[tool.ruff.lint.flake8-annotations]
+mypy-init-return = true
+
+[tool.ruff.format]
+docstring-code-format = true
diff --git a/src/ga4gh/core/__init__.py b/src/ga4gh/core/__init__.py
@@ -1,47 +1,51 @@
-"""Python support used across GA4GH projects
+"""Python support used across GA4GH projects"""
 
-"""
+from importlib.metadata import PackageNotFoundError, version
 
-from importlib.metadata import version, PackageNotFoundError
-
-from .digests import sha512t24u
-from .enderef import ga4gh_enref, ga4gh_deref
-from .identifiers import (
-    ga4gh_digest, ga4gh_identify, ga4gh_serialize, is_ga4gh_identifier,
-    VrsObjectIdentifierIs, use_ga4gh_compute_identifier_when,
-    CURIE_NAMESPACE, CURIE_SEP, GA4GH_PREFIX_SEP, GA4GH_IR_REGEXP, GA4GH_DIGEST_REGEXP,
-    PrevVrsVersion
-)
-from .pydantic import (
-    is_pydantic_instance, is_curie_type, pydantic_copy
+import ga4gh.core.models as core_models
+from ga4gh.core.digests import sha512t24u
+from ga4gh.core.enderef import ga4gh_deref, ga4gh_enref
+from ga4gh.core.identifiers import (
+    CURIE_NAMESPACE,
+    CURIE_SEP,
+    GA4GH_DIGEST_REGEXP,
+    GA4GH_IR_REGEXP,
+    GA4GH_PREFIX_SEP,
+    PrevVrsVersion,
+    VrsObjectIdentifierIs,
+    ga4gh_digest,
+    ga4gh_identify,
+    ga4gh_serialize,
+    is_ga4gh_identifier,
+    use_ga4gh_compute_identifier_when,
 )
-from . import models as core_models
+from ga4gh.core.pydantic import is_curie_type, is_pydantic_instance, pydantic_copy
 
 __all__ = [
-    "sha512t24u",
-    "ga4gh_enref",
+    "CURIE_NAMESPACE",
+    "CURIE_SEP",
+    "GA4GH_DIGEST_REGEXP",
+    "GA4GH_IR_REGEXP",
+    "GA4GH_PREFIX_SEP",
+    "PrevVrsVersion",
+    "VrsObjectIdentifierIs",
+    "core_models",
     "ga4gh_deref",
     "ga4gh_digest",
+    "ga4gh_enref",
     "ga4gh_identify",
     "ga4gh_serialize",
+    "is_curie_type",
     "is_ga4gh_identifier",
-    "VrsObjectIdentifierIs",
-    "use_ga4gh_compute_identifier_when",
-    "CURIE_NAMESPACE",
-    "CURIE_SEP",
-    "GA4GH_PREFIX_SEP",
-    "GA4GH_IR_REGEXP",
-    "GA4GH_DIGEST_REGEXP",
-    "PrevVrsVersion",
     "is_pydantic_instance",
-    "is_curie_type",
     "pydantic_copy",
-    "core_models",
+    "sha512t24u",
+    "use_ga4gh_compute_identifier_when",
 ]
 
 try:
     __version__ = version(__name__)
-except PackageNotFoundError:    # pragma: nocover
+except PackageNotFoundError:  # pragma: nocover
     __version__ = "unknown"
 finally:
     del version, PackageNotFoundError
diff --git a/src/ga4gh/core/digests.py b/src/ga4gh/core/digests.py
@@ -9,8 +9,8 @@
 import hashlib
 
 
-def sha512t24u(blob):
-    """generate a base64url-encode, truncated SHA-512 digest for given
+def sha512t24u(blob: bytes) -> str:
+    """Generate a base64url-encode, truncated SHA-512 digest for given
     binary data
 
     The sha512t24u digest is a convention for constructing and
@@ -21,14 +21,13 @@ def sha512t24u(blob):
         * encode using base64url encoding
 
     Examples:
-    >>> sha512t24u(b'')
+    >>> sha512t24u(b"")
     'z4PhNX7vuL3xVChQ1m2AB9Yg5AULVxXc'
 
     >>> sha512t24u(b"ACGT")
     'aKF498dAxcJAqme6QYQ7EZ07-fiw8Kw2'
 
     """
-
     digest_size = 24
     digest = hashlib.sha512(blob).digest()
     tdigest_b64us = base64.urlsafe_b64encode(digest[:digest_size])